Differential D32984 Diff 98477 sys/dev/random/randomdev.c

Changeset View

Standalone View

sys/dev/random/randomdev.c

Show First 20 Lines • Show All 334 Lines • ▼ Show 20 Lines	randomdev_write(struct cdev dev __unused, struct uio uio, int flags __unused)
random_buf = malloc(PAGE_SIZE, M_ENTROPY, M_WAITOK);		random_buf = malloc(PAGE_SIZE, M_ENTROPY, M_WAITOK);
nbytes = uio->uio_resid;		nbytes = uio->uio_resid;
while (uio->uio_resid > 0 && error == 0) {		while (uio->uio_resid > 0 && error == 0) {
c = MIN(uio->uio_resid, PAGE_SIZE);		c = MIN(uio->uio_resid, PAGE_SIZE);
error = uiomove(random_buf, c, uio);		error = uiomove(random_buf, c, uio);
if (error)		if (error)
break;		break;
randomdev_accumulate(random_buf, c);		randomdev_accumulate(random_buf, c);
tsleep(p_random_alg_context, 0, "randwr", hz/10);
}		}
if (nbytes != uio->uio_resid && (error == ERESTART \|\| error == EINTR))		if (nbytes != uio->uio_resid && (error == ERESTART \|\| error == EINTR))
/* Partial write, not error. */		/* Partial write, not error. */
error = 0;		error = 0;
		cemUnsubmitted Not Done Inline Actions Is this condition even possible anymore? The only way I can see the accumulation loop exiting early is if we hit trap on an invalid userspace buffer, producing `EFAULT`. `randomdev_accumulate()` is incapable of failure. cem: Is this condition even possible anymore? The only way I can see the accumulation loop exiting…
		cpercivaAuthorUnsubmitted Done Inline Actions Maybe possible if there's a page fault and a signal arrives while the data is being faulted in? I agree that it's unlikely though. cperciva: Maybe possible if there's a page fault and a signal arrives while the data is being faulted in?
		cemUnsubmitted Not Done Inline Actions If that case does happen, should we squash actually error, rather than returning EINTR? I thought we were supposed to bubble that up to the syscall entry layer, generally. Anyway, orthogonal to your improvement here. Thanks! cem: If that case does happen, should we squash actually error, rather than returning EINTR? I…
		cpercivaAuthorUnsubmitted Done Inline Actions No, this code is correct -- we return EINTR from the syscall if the write was interrupted before writing anything (i.e. if `uio_resid` is still equal to `nbytes`), but if at least one byte was written then write(2) returns success with a short write length. cperciva: No, this code is correct -- we return EINTR from the syscall if the write was interrupted…
		cemUnsubmitted Not Done Inline Actions Doesn’t dofilewrite already do that? http://fxr.watson.org/fxr/source/kern/sys_generic.c#L568 cem: Doesn’t dofilewrite already do that? http://fxr.watson.org/fxr/source/kern/sys_generic.c#L568
		cpercivaAuthorUnsubmitted Done Inline Actions Hmm yes, indeed it seems that way. I suspect this isn't the only place this sort of duplication is present in the tree, either. cperciva: Hmm yes, indeed it seems that way. I suspect this isn't the only place this sort of…
free(random_buf, M_ENTROPY);		free(random_buf, M_ENTROPY);
return (error);		return (error);
}		}

/* ARGSUSED */		/* ARGSUSED */
static int		static int
randomdev_poll(struct cdev dev __unused, int events, struct thread td __unused)		randomdev_poll(struct cdev dev __unused, int events, struct thread td __unused)
{		{
▲ Show 20 Lines • Show All 78 Lines • Show Last 20 Lines